CN115130002A

CN115130002A - Recommendation request processing method and device, computer equipment and storage medium

Info

Publication number: CN115130002A
Application number: CN202210870465.7A
Authority: CN
Inventors: 张方方; 汤浪
Original assignee: Shanghai Yuer Network Technology Co ltd
Current assignee: Shanghai Yuer Network Technology Co ltd
Priority date: 2022-07-22
Filing date: 2022-07-22
Publication date: 2022-09-30

Abstract

The application relates to a recommendation request processing method, a recommendation request processing device, a recommendation request processing computer device, a storage medium and a recommendation request processing computer program product. The recommendation request processing method comprises the following steps: receiving a recommendation request, wherein the recommendation request carries a scene identifier; acquiring configuration information corresponding to the scene identifier; generating corresponding recommended service according to the configuration information; and obtaining a recommendation result according to the recommendation service. A recommendation service configuration method, comprising: receiving a configured scene identifier; receiving configuration information for the scene identification; storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plug-ins of the recommendation links and parameters corresponding to the plug-ins. By adopting the two methods, the development efficiency of the recommendation service can be improved.

Description

Recommendation request processing method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of information technologies, and in particular, to a recommendation request processing method, apparatus, computer device, storage medium, and computer program product.

Background

With the development of information technology, users gradually move from the information-deficient era to the information-overloaded era, so that much information never gets the attention of the users and disappears in the information ocean. And the recommendation system makes the information recommended to the user stand out from the overloaded massive information, and the user can pay attention to the information.

The development project of the existing recommendation system faces different recommendation service scenes, a large number of similar codes need to be repeatedly written to achieve different recommendation requirements, and the development efficiency of the development project of the recommendation system is low.

Disclosure of Invention

In view of the above, it is necessary to provide a recommendation request processing method, apparatus, computer device, computer readable storage medium, and computer program product capable of improving the development efficiency of recommendation services in view of the above technical problems.

In a first aspect, the present application provides a recommendation request processing method, including: receiving a recommendation request, wherein the recommendation request carries a scene identifier; acquiring configuration information corresponding to the scene identifier; generating corresponding recommended service according to the configuration information; and obtaining a recommendation result according to the recommendation service.

In one embodiment, the obtaining the configuration information corresponding to the scene identifier includes: calling a configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a database; determining a configuration plug-in corresponding to each recommended link from the database; determining configuration parameters corresponding to the configuration plug-ins from the database; and outputting the recommendation link, the configuration plug-in and the configuration parameters as configuration information.

In one embodiment, the invoking the configuration query service includes: calling the configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a cache; if the cache comprises the recommendation links corresponding to the scene identification, determining configuration plug-ins corresponding to the recommendation links from the cache; determining configuration parameters corresponding to the configuration plug-ins from the cache; if the recommendation link corresponding to the scene identification is not included in the cache, acquiring the recommendation link corresponding to the scene identification from the database; determining configuration plug-ins corresponding to the recommended links from the database; and determining configuration parameters corresponding to each configuration plug-in from the database.

In one embodiment, the recommendation request carries a user identifier; the obtaining of the configuration information corresponding to the scene identifier includes: determining configuration conditions corresponding to the user identification from at least two groups of configuration conditions of the recommended link; and determining a configuration plug-in corresponding to the recommendation link according to the configuration condition.

In one embodiment, after obtaining the recommendation result according to the recommendation service, the method further includes: acquiring a log of a recommended service, wherein the log comprises a recommended result returned according to the recommended service; analyzing the recommendation result in each log to obtain a log analysis result, and adjusting the configuration information according to the log analysis result; and performing recommended service debugging based on the adjusted configuration information.

In one embodiment, the obtaining of the recommendation result according to the recommendation service includes: acquiring a current plug-in a current recommendation link; executing the current plug-in to obtain a current intermediate processing result; and obtaining a next plugin in the next recommendation link, taking the next plugin as a current plugin, obtaining a next intermediate processing result based on the current intermediate processing result and the current plugin, taking the next intermediate processing result as the current intermediate processing result, and continuously executing the step of obtaining the next plugin in the next recommendation link until all recommendation links are executed, so as to obtain a recommendation result.

In one embodiment, the executing the current plug-in unit to obtain the current intermediate processing result includes: and if the number of the current plug-ins is more than one, performing serial processing or parallel processing on each current plug-in according to the preset identification of each current plug-in to obtain the current intermediate processing result.

In a second aspect, the present application further provides a recommendation service configuration method, where the recommendation service configuration method includes: receiving a configured scene identifier; receiving configuration information for the scene identity; storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plug-ins of the recommendation links and parameters corresponding to the plug-ins.

In one embodiment, the receiving the configuration information for the scene identification includes: receiving at least two sets of configuration conditions of at least one of the recommended links; and storing at least two groups of configuration conditions in association with the recommended links.

In a third aspect, the present application further provides a recommendation request processing apparatus, including: the system comprises a receiving module, a recommending module and a judging module, wherein the receiving module is used for receiving a recommending request, and the recommending request carries a scene identifier; an obtaining module, configured to obtain configuration information corresponding to the scene identifier; the generating module is used for generating corresponding recommended services according to the configuration information; and the recommendation result obtaining module is used for obtaining a recommendation result according to the recommendation service.

In a fourth aspect, the present application further provides a recommendation service configuration apparatus, where the recommendation service configuration apparatus includes: the identification receiving module is used for receiving the configured scene identification; a configuration receiving module, configured to receive configuration information for the scene identifier; the storage module is used for storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plug-ins of the recommendation links and parameters corresponding to the plug-ins.

In a fifth aspect, the application further provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of the method when executing the computer program.

In a sixth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method.

In a seventh aspect, the present application further provides a computer program product. The computer program product comprises a computer program which, when being executed by a processor, carries out the steps of the method.

According to the information recommendation configuration method, the information recommendation configuration device, the computer equipment, the storage medium and the computer program product, the configuration information corresponding to the scene identification is obtained through the scene identification carried by the recommendation request, the corresponding recommendation service is generated according to the configuration information, and the recommendation result aiming at the recommendation request is obtained according to the recommendation service. Different scene identifications correspond to different configuration information, so that different recommendation services can be generated according to different configuration information. Different recommendation services can be generated through different configuration information, universal codes corresponding to the configuration information do not need to be repeatedly written, and development efficiency of the recommendation services is improved. Different recommendation services generated by the configuration information can increase the white-box of the recommendation services and facilitate the realization and maintenance of the recommendation services by developers.

Drawings

FIG. 1 is a diagram of an application environment for a method of processing a recommendation request in one embodiment;

FIG. 2 is a flowchart illustrating a method for processing a recommendation request in one embodiment;

FIG. 3 is a flow diagram that illustrates a method for recommending service configurations in an embodiment;

FIG. 4 is a block diagram illustrating a method for processing a recommendation request in accordance with an embodiment;

FIG. 5 is a schematic diagram of a front end of a recommended service configuration method in an embodiment;

FIG. 6 is a schematic diagram of a front end of a recommended service configuration method in one embodiment;

FIG. 7 is a schematic diagram of a front end of a recommended service configuration method in one embodiment;

FIG. 8 is a schematic diagram of a front end of a recommended service configuration method in one embodiment;

FIG. 9 is a schematic diagram of a front end of a recommended service configuration method in an embodiment;

FIG. 10 is a schematic diagram of a front end of a recommended service configuration method in one embodiment;

FIG. 11 is a flowchart illustrating step 208 in one embodiment;

FIG. 12 is a block diagram showing a configuration of a recommendation request processing apparatus in one embodiment;

FIG. 13 is a block diagram showing the configuration of a recommendation service configuration apparatus in one embodiment;

FIG. 14 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The recommendation request processing method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104, or may be located on the cloud or other network server. The terminal 102 sends a recommendation request to the server 104. The server 104 is configured to receive a recommendation request, where the recommendation request carries a scene identifier; acquiring configuration information corresponding to the scene identifier; generating corresponding recommended service according to the configuration information; and obtaining a recommendation result according to the recommendation service. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices and portable wearable devices, and the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart car-mounted devices, and the like. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, and the like. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers.

In one embodiment, as shown in fig. 2, a recommendation request processing method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:

step 202, a recommendation request is received, where the recommendation request carries a scene identifier.

The recommendation request is a request for acquiring recommendation information triggered according to the behavior action of the user, and the server provides the corresponding recommendation information for the user through the recommendation request. It should be noted that the present application does not limit the behavior action of triggering the recommendation request by the user, and may trigger the recommendation request and send the recommendation request to the server, including but not limited to receiving a mouse click preset buried point triggering recommendation request and receiving a search button control triggering recommendation request. The scene identifier represents a scene where the user side triggers the recommendation request, and it can be understood that the scene identifiers corresponding to different scenes are different.

Optionally, when the user clicks the interface to trigger the recommendation request, the terminal where the interface is located sends the recommendation request to the server, where the recommendation request includes a scene identifier triggering the recommendation request.

And the server receives a recommendation request carrying the scene identifier sent by the terminal.

Step 204, obtaining configuration information corresponding to the scene identifier.

The configuration information comprises recommendation links for generating recommendation services, configuration plugins called by the recommendation links and specific configuration parameters when the configuration plugins are executed. It should be noted that, recommendation requests in different scenarios have different recommendation information to be returned, and therefore, the configuration information of generating recommendation services corresponding to different scenarios is different. It should be noted that the configuration information in each scene is preset, and the scene identifier of each scene is stored in association with the corresponding configuration information.

The server acquires corresponding configuration information according to the scene identification in the recommendation request, wherein the configuration information comprises recommendation links for generating recommendation services corresponding to the recommendation request, configuration plug-ins called by the recommendation links and specific configuration parameters when the configuration plug-ins are executed.

And step 206, generating a corresponding recommended service according to the configuration information.

And the recommendation service is a service thread which acquires recommendation information according to the recommendation request.

And the server generates corresponding recommendation service according to the configuration information corresponding to the recommendation request.

And step 208, obtaining a recommendation result according to the recommendation service.

And the server acquires recommendation information, namely a recommendation result, corresponding to the recommendation request according to the recommendation service and sends the recommendation result to the terminal.

In the recommendation request processing method, the configuration information corresponding to the scene identifier is obtained through the scene identifier carried by the recommendation request, the corresponding recommendation service is generated according to the configuration information, and the recommendation result aiming at the recommendation request is obtained according to the recommendation service. Different scene identifications correspond to different configuration information, so that different recommendation services can be generated according to different configuration information. Different recommendation services can be generated through different configuration information, universal codes corresponding to the configuration information do not need to be repeatedly written, and development efficiency of development projects of the recommendation services is improved. Different recommendation services generated by the configuration information can increase the white-box of the recommendation services and facilitate the realization and maintenance of the recommendation services by developers.

In one embodiment, obtaining configuration information corresponding to a scene identifier includes: calling configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a database; determining configuration plug-ins corresponding to the recommended links from a database; determining configuration parameters corresponding to the configuration plug-ins from a database; and outputting the recommendation link, the configuration plug-in and the configuration parameters as configuration information.

The configuration query service is used for searching configuration information corresponding to the scene identifier, and the configuration information comprises a specific recommendation link, configuration plugins called by the recommendation link and configuration parameters executed by the configuration plugins. The recommendation link is an intermediate link for executing the recommendation service to obtain a recommendation result, and the recommendation link includes but is not limited to preprocessing, recalling, filling, filtering, merging, sorting, reordering, caching, rule processing and return value processing links. It should be noted that recommendation links and sequences in configuration information corresponding to different scene identifiers may be different. There are a number of types of configuration plug-ins that can be invoked by each recommendation link. If one configuration link calls a plurality of configuration plugins, the types of the called configuration plugins can be the same or different. The configuration parameters are execution parameters of the configuration plug-in, and for example, the recalled configuration parameters include, but are not limited to, an attribute, a recall address, and a recall number of the candidate object to be recalled. It should be noted that the association relationship between the scene identifier and the corresponding configuration information is a preset association relationship. Optionally, the configuration information corresponding to the scene identifier is stored in a database.

Specifically, the server calls a configuration query service to acquire a recommendation link corresponding to the scene identifier from a database; determining configuration plug-ins corresponding to the recommended links from a database; determining configuration parameters corresponding to the configuration plug-ins from a database; and outputting the recommendation link, the configuration plug-in and the configuration parameters as configuration information.

In the recommendation request processing method, the configuration information corresponding to the scene identifier can be quickly acquired through the configuration query service.

In one embodiment, invoking a configuration query service includes: calling a configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a cache; if the cache comprises recommendation links corresponding to the scene identification, determining configuration plug-ins corresponding to the recommendation links from the cache; determining configuration parameters corresponding to the configuration plug-ins from the cache; if the cache does not comprise a recommendation link corresponding to the scene identifier, acquiring the recommendation link corresponding to the scene identifier from the database; determining configuration plug-ins corresponding to the recommended links from a database; and determining the configuration parameters corresponding to the configuration plug-ins from the database.

Optionally, one end of the cache is connected to the configuration query service, and the other end of the cache is connected to the database. The cache is used for storing the configuration parameters corresponding to the scene identifiers.

Specifically, the server calls a configuration query service, and acquires a recommendation link corresponding to the scene identifier from a cache; if the cache comprises recommendation links corresponding to the scene identification, determining configuration plug-ins corresponding to the recommendation links from the cache; determining configuration parameters corresponding to the configuration plug-ins from the cache; if the cache does not comprise a recommendation link corresponding to the scene identifier, acquiring the recommendation link corresponding to the scene identifier from the database; determining configuration plug-ins corresponding to the recommended links from a database; and determining configuration parameters corresponding to each configuration plug-in from a database, and loading the configuration information corresponding to the scene identifier acquired from the database into a cache.

In the recommendation request processing method, the configuration parameter data corresponding to the recently used scene identifier is cached and stored, so that the speed of acquiring the configuration information corresponding to the scene identifier is improved.

In one embodiment, the recommendation request carries a user identification; acquiring configuration information corresponding to the scene identifier, including: determining configuration conditions corresponding to the user identification from at least two groups of configuration conditions of the recommendation link; and determining a configuration plug-in corresponding to the recommendation link according to the configuration condition.

The user identification is used for representing the user triggering the recommendation request, and the user identifications corresponding to different users are different. It should be noted that the recommendation link in the configuration information corresponding to the scene identifier may include two sets of configuration conditions, optionally, one set of configuration conditions is an experimental set of configuration conditions, and one set of configuration conditions is a comparison set of configuration conditions. If the number of groups of configuration conditions of the recommended links is greater than 2, a plurality of configuration conditions of the control group can be provided, and a plurality of configuration conditions of the experimental group can be provided. The configuration plug-ins corresponding to the recommendation links and the configuration parameters corresponding to the configuration plug-ins are collectively referred to as a set of configuration conditions of one recommendation link. When a comparison experiment is needed, at least two sets of configuration conditions may be configured for one recommendation link, but the generation of the recommendation service only needs one set of configuration conditions of one recommendation link, so that the server determines one set of recommendation conditions from the at least two sets of recommendation conditions of the recommendation link as the configuration conditions of the recommendation link according to the user identification.

The recommendation request received by the server includes a user identification. The server acquires at least two sets of configuration conditions of the recommendation link corresponding to the scene identification, and determines one set of recommendation conditions from the at least two sets of recommendation conditions of the recommendation link as the configuration conditions of the recommendation link according to the user identification. And the server determines the configuration plug-in corresponding to the recommendation link and the configuration parameters corresponding to the configuration plug-in according to the determined configuration conditions. Optionally, the server is implemented by a test service, and determines, according to the user identifier, one of the at least two sets of recommendation conditions in the recommendation link as a determination algorithm of the configuration condition of the recommendation link.

In the recommendation request processing method, the server determines the configuration conditions corresponding to the user identification from at least two groups of configuration conditions of each recommendation link, and generates the corresponding recommendation service. The server side is used for shunting the users of the recommended services of the experimental group and the comparison group, and developers are not required to test codes of the recommended services to generate a plurality of recommended services.

In one embodiment, after obtaining the recommendation result according to the recommendation service, the method further includes: acquiring a log of the recommended service, wherein the log comprises a recommendation result returned according to the recommended service; analyzing the recommended result in each log to obtain a log analysis result, and adjusting the configuration information according to the log analysis result; and performing recommended service debugging based on the adjusted configuration information.

In one embodiment, obtaining the recommendation result according to the recommendation service includes: acquiring a current plug-in a current recommendation link; executing the current plug-in to obtain a current intermediate processing result; and acquiring a next plugin in the next recommendation link, taking the next plugin as the current plugin, acquiring a next intermediate processing result based on the current intermediate processing result and the current plugin, taking the next intermediate processing result as the current intermediate processing result, and continuously executing the step of acquiring the next plugin in the next recommendation link until all recommendation links are executed to obtain the recommendation result.

The recommendation service comprises at least one recommendation link, wherein the recommendation link comprises but is not limited to preprocessing, recalling, filling, filtering, merging, sorting, reordering, caching, rule processing and return value processing links. The following description will be given by taking the links of recommendation of the recommendation service including preprocessing, recall, padding, filtering, merging, sorting, reordering, caching, rule processing, and return value processing as examples.

The server sets the preprocessing as a current recommendation link and acquires a first plug-in corresponding to the preprocessing; and executing the first plug-in to obtain a first candidate object, and taking the first candidate object as a current intermediate processing result. The first plug-in is used for loading first candidate objects needing filtering from different dimensions, wherein the first candidate objects include but are not limited to items recommended to a current user (namely, a user triggering a recommendation request) within a preset time range, and items exposed and/or clicked within the preset time range, and the number of times of the items is larger than a preset value. And the server acquires a next plugin in the next recommendation link, namely a second plugin, takes the second plugin as a current plugin, acquires a next intermediate processing result, namely a second candidate object, based on the current intermediate processing result (the first candidate object) and the current plugin (the second plugin), takes the second candidate object as the current intermediate processing result, and continues to execute the step of acquiring the next plugin in the next recommendation link, namely a third plugin until the execution of all recommendation links is completed to acquire the recommendation result. It should be noted that the second plug-in is used to recall the initial candidate object, and the process of obtaining the second candidate object based on the first candidate object and the second plug-in includes filtering a portion of the initial candidate object that overlaps with the first candidate object to obtain the second candidate object.

In one embodiment, executing the current plug-in to obtain the current intermediate processing result includes: and if the number of the current plug-ins is more than one, performing serial processing or parallel processing on each current plug-in according to the preset identification of each current plug-in to obtain a current intermediate processing result.

The preset identifier is used for indicating that the processing mode of the current plug-in is serial or parallel, and if the processing mode of the current plug-in is serial, the preset identifier is also used for indicating the processing sequence of the current plug-in.

In one embodiment, as shown in fig. 3, there is provided a recommendation service configuration method, including:

step 302, receiving a configured scene identifier.

The server receives a scene identification configured by a user.

Step 304, receiving configuration information for the scene identification.

It should be noted that different scene identifiers correspond to different recommendation services, and configuration information of each recommendation service is different.

The server receives configuration information aiming at the scene identification, wherein the configuration information comprises but is not limited to recommendation links, configuration plugins called by the recommendation links and specific configuration parameters when the configuration plugins are executed.

Step 306, storing the scene identification and the configuration information in a correlation manner; the configuration information comprises recommendation links, plugins of the recommendation links and parameters corresponding to the plugins.

The server stores the scene identification and the corresponding configuration information in an associated manner; the configuration information comprises recommendation links, plugins of the recommendation links and parameters corresponding to the plugins. Optionally, the configuration information corresponding to the scene identification is stored in a database.

According to the recommendation service configuration method, the server receives different configuration information corresponding to different scene identifiers, different recommendation services can be generated according to the different scene identifiers, different recommendation services can be generated through the different configuration information, complex codes do not need to be repeatedly written according to the different recommendation services, and the configuration information is white-boxed, so that the development process is transparent and easy to maintain.

In one embodiment, receiving configuration information for a scene identification comprises: receiving at least two groups of configuration conditions of at least one recommended link; and storing at least two groups of configuration conditions in association with the recommended links.

It should be noted that the server generates a corresponding recommendation service according to a set of recommendation conditions of a recommendation link corresponding to the scene identifier.

In the recommendation service configuration method, the server configures the configuration conditions of the experimental group and the comparison group of each recommendation link by setting a plurality of groups of configuration conditions for the recommendation links, only the configuration conditions of the experimental group and the comparison group of each recommendation link need to be considered, how to match the experimental group and the comparison group of each recommendation link with the corresponding users does not need to be considered, the configuration conditions of the experimental group and the comparison group of each recommendation link can be added, and the method is flexible.

In a specific embodiment, as shown in fig. 4, the server receives the scene identifier and the configuration information corresponding to the scene identifier through a management-service (management service), and stores the configuration information corresponding to the scene identifier in a database (MySQL). And the User triggers a recommendation request and sends the recommendation request to the server. The server receives a recommendation request which is sent by a user and carries a scene identifier and a user identifier as shown by a connecting line 1, sends the recommendation request to a Backend-Service (Backend Service), sends a RecEngine (recommendation Engine SDK) with the recommendation request as shown by a connecting line 2, encapsulates the scene identifier and the user identifier in a context object through the RecEngine, and sends the context object to a Config-Service (recommendation configuration analysis Service), namely, invokes a configuration query Service. After the Config-Service analyzes the scene identifier in the context object, as shown by the connecting line 4, the Config-Service acquires the configuration information corresponding to the scene identifier from the Redis cache through the connecting line 4. If the configuration information corresponding to the scene identifier does not exist in the cache, the Config-Service acquires the configuration information corresponding to the scene identifier from the MySQL (database), and stores the configuration information corresponding to the scene identifier acquired from the database into the cache. The Config-Service judges the configuration information corresponding to the scene identification, and if the configuration condition corresponding to any one of the recommendation links in the configuration information is greater than 1, that is, the recommendation link comprises at least two groups of configuration conditions, the configuration conditions corresponding to the recommendation link and the user identification in the context object are sent to the Ab-Test-Service (AB experimental platform) through a connecting line 6. And the Ab-Test-Service determines one group of configuration conditions from at least two groups of configuration conditions of the recommended link according to a built-in algorithm and the user identification as final configuration conditions of the recommended link, stores the association relationship between the final configuration conditions and the user identification, and returns the user identification, namely the corresponding final configuration conditions, to the Config-Service as shown by a connecting line 7. And the Config-Service returns the final configuration information corresponding to the scene identification to the RecEngine. The RecEngine generates corresponding recommendation service according to the final configuration information; and obtaining a recommendation result according to the recommendation service, and returning the recommendation result to the user.

Optionally, a log-collect-client (log collection end) acquires a log of the recommended service, where the log includes the recommended result returned by the recommended service, and the log of the recommended service is sent to Es (database). The method comprises the steps that a manager-service acquires logs of recommended services from Es, analyzes recommendation results in the logs to obtain log analysis results, and adjusts configuration information according to the log analysis results; and performing recommended service debugging based on the adjusted configuration information.

Optionally, the manager-service may simulate a recommendation request triggered by a user, directly send the simulated recommendation request to the RecEngine, and perform a step of encapsulating the scene identifier and the user identifier in the simulation request in the context object by the RecEngine, so as to implement a test on the recommendation service.

Optionally, the server receives the scene identifier and the implementation of the configuration information corresponding to the scene identifier through a management-service (management service), which is implemented by a front-end page as shown in fig. 5. Scene (Scene name) in fig. 5 is a Scene identifier, and the PreFiller (preprocessing link), Loader (recall), Filler (fill), Filter (Filter), merge (merge), Ranker (sort), ReRanker (reorder), Cache (Cache), ResultSelector (rule process), ResponseFiller (return value process), and ExtParameter (timer) in fig. 5 are options for receiving configuration plug-ins of each recommended link and configuration parameters corresponding to each configuration plug-in. It should be noted that the recommendation link corresponding to the scene identifier may include, but is not limited to, the recommendation link in fig. 5, and may also define other recommendation links by user, and in addition, in the process of configuring the recommendation link corresponding to the scene identifier, only a part of the recommendation link in fig. 5 may be used, and the order of the recommendation link in fig. 5 may be adaptively changed.

Alternatively, the user may configure a plurality of configuration plug-ins for one recommendation link, as shown in fig. 6, for example, two plug-ins for recall. And respectively configuring the names of two recalling plug-ins on a Loader configuration interface, wherein the name of the first recalling plug-in is redISLoader, and the name of the second recalling plug-in is EsLoader. As shown in fig. 7, the configuration parameters of the first recall plug-in are configured, where the storage structure (RedisDataType) is a zset structure, value is the primary key id of candidate, and score is the score corresponding to candidate. When using the RedisLoader plug-in, a developer can obtain the result of the recall by only configuring the key of the redis (for the instance name of the redis (DbNamespace in FIG. 6), the default configuration can be provided in sdk, and meanwhile, the corresponding instance can be obtained on the page according to the transmitted configuration analysis). The configuration of the configuration parameters of the second recall plug-in is shown in fig. 8, and a developer only needs to configure a simple query statement, an index name of es, an index type, and a primary key field name of candidate in es, so as to obtain a recall result. In the process that RecEngine obtains a recommendation result through a recommendation service, an es recall plug-in (a second recall plug-in) is used for packaging the analysis of an es query statement, mapping the query result to a candidate class, automatically filling the _ source content in the query result into a dynamic property set of the candidate, filling the _ score of the es into the score of the candidate and finishing the processing of various exceptions.

Optionally, as shown in fig. 9, the sequencing stage is configured, the configuration plug-in configured in the sequencing stage is a rank-service (an online sequencing service, which implements a function of model sequencing), the rank-service provides a dubbo interface, and the configuration parameters corresponding to the configuration plug-in include: the model name, the model type, the model interface name and the configuration plug-in can realize the personalized sequencing of the candidate set list after the configuration parameters are configured. In the process that RecEngine obtains a recommendation result through a recommendation service, a ranking stage completes the request of rank-service in a dubbo generalized call-based mode, obtains the real-time model scores of candidate objects of each candidate object to the current requesting user, and completes ranking based on the scores.

Optionally, as shown in fig. 10, the reordering stage is configured, and the configuration condition and the returned score input by the developer are received. In the process that the RecEngine obtains the recommendation result through the recommendation service, the reordering stage analyzes the configuration conditions to obtain score, and finishes the ordering according to the candidate final score.

The flow of the server obtaining the recommendation result according to the recommendation service is shown in fig. 11.

It should be noted that the configuration plug-ins configurable in each recommendation link are general plug-ins abstracted in advance according to each technology stack.

Optionally, the candidate set object attribute and the request context object attribute are both defined as currentHashMap class to support dynamic parameter passing. The context object is used for storing the scene identification and the user identification of the recommendation request. The candidate set object attribute is used for storing the attribute of the candidate object, and the candidate object is an intermediate result of the recommendation result.

It should be understood that, although the steps in the service diagram related to the embodiments described above are sequentially displayed as indicated by the arrows, the steps are not necessarily sequentially executed in the order indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the service diagram according to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the present application further provides a recommendation request processing apparatus for implementing the recommendation request processing method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the method, so the specific limitations in one or more embodiments of the recommendation request processing device provided below may refer to the limitations on the recommendation request processing method in the foregoing, and details are not described here.

In one embodiment, as shown in fig. 12, there is provided a recommendation request processing apparatus including: the recommendation system comprises a receiving module 100, an obtaining module 200, a generating module 300 and a recommendation result obtaining module 400, wherein: the receiving module 100 is configured to receive a recommendation request, where the recommendation request carries a scene identifier. An obtaining module 200, configured to obtain configuration information corresponding to the scene identifier. A generating module 300, configured to generate a corresponding recommended service according to the configuration information. A recommendation result obtaining module 400, configured to obtain a recommendation result according to the recommendation service.

In one embodiment, the obtaining module includes: the first calling module is used for calling the configuration query service and acquiring a recommendation link corresponding to the scene identifier from the database; the first determining module is used for determining the configuration plug-ins corresponding to the recommending links from the database; the second determining module is used for determining the configuration parameters corresponding to the configuration plug-ins from the database; and the output module is used for outputting the recommended links, the configuration plug-ins and the configuration parameters as configuration information.

In one embodiment, the first calling module includes: the second calling module is used for calling the configuration query service and acquiring a recommendation link corresponding to the scene identifier from the cache; the third determining module is used for determining the configuration plug-ins corresponding to the recommending links from the cache if the recommending links corresponding to the scene identification are included in the cache; determining configuration parameters corresponding to the configuration plug-ins from the cache; the fourth determining module is used for acquiring the recommending link corresponding to the scene identifier from the database if the recommending link corresponding to the scene identifier is not included in the cache; determining configuration plug-ins corresponding to the recommended links from a database; and determining the configuration parameters corresponding to the configuration plug-ins from the database.

In one embodiment, an acquisition module includes: the fifth determining module is used for determining the configuration conditions corresponding to the user identification from at least two groups of configuration conditions of the recommended link; and the sixth determining module is used for determining the configuration plug-ins corresponding to the recommending links according to the configuration conditions.

In one embodiment, further comprising: the log obtaining module is used for obtaining a log of the recommended service, wherein the log comprises a recommendation result returned according to the recommended service; the log analysis module is used for analyzing the recommendation result in each log to obtain a log analysis result and adjusting the configuration information according to the log analysis result; and the debugging module is used for carrying out recommendation service debugging based on the adjusted configuration information.

In one embodiment, the recommendation obtaining module includes: the plug-in acquisition module is used for acquiring a current plug-in a current recommendation link; the execution module is used for executing the current plug-in to obtain a current intermediate processing result; and the circulation module is used for acquiring a next plugin in the next recommendation link, taking the next plugin as the current plugin, acquiring a next intermediate processing result based on the current intermediate processing result and the current plugin, taking the next intermediate processing result as the current intermediate processing result, and continuously executing the step of acquiring the next plugin in the next recommendation link until all recommendation links are executed, so as to acquire the recommendation result.

In one embodiment, an execution module includes: and the processing module is used for performing serial processing or parallel processing on each current plug-in according to the preset identification of each current plug-in if the number of the current plug-ins is more than one to obtain a current intermediate processing result.

Based on the same inventive concept, the embodiment of the present application further provides a recommendation service configuration device for implementing the above-mentioned recommendation service configuration method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so specific limitations in one or more embodiments of the recommendation service configuration device provided below can be referred to the limitations of the recommendation service configuration method in the foregoing, and details are not described herein again.

In one embodiment, as shown in fig. 13, there is provided a recommendation service configuration apparatus including: an identity receiving module 500, a configuration receiving module 600, and a storage module 700, wherein: the identification receiving module is used for receiving the configured scene identification; a configuration receiving module, configured to receive configuration information for the scene identifier; the storage module is used for storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plugins of the recommendation links and parameters corresponding to the plugins.

In one embodiment, the identity reception module comprises: the first receiving module is used for receiving at least two groups of configuration conditions of at least one recommended link; and the association storage module is used for associating and storing at least two groups of configuration conditions and the recommendation links.

The respective modules in the recommendation request processing apparatus and the recommendation service configuration described above may be implemented in whole or in part by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 14. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a recommendation request processing method and a recommendation service configuration method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 14 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program: receiving a recommendation request, wherein the recommendation request carries a scene identifier; acquiring configuration information corresponding to the scene identifier; generating corresponding recommended service according to the configuration information; and obtaining a recommendation result according to the recommendation service.

In one embodiment, the obtaining configuration information corresponding to the scene identifier, which is implemented when the processor executes the computer program, includes: calling configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a database; determining configuration plug-ins corresponding to the recommended links from a database; determining configuration parameters corresponding to the configuration plug-ins from a database; and outputting the recommendation link, the configuration plug-in and the configuration parameters as configuration information.

In one embodiment, invoking a configuration query service, implemented by a processor executing a computer program, comprises: calling a configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a cache; if the cache comprises recommendation links corresponding to the scene identification, determining configuration plug-ins corresponding to the recommendation links from the cache; determining configuration parameters corresponding to the configuration plug-ins from the cache; if the cache does not comprise a recommendation link corresponding to the scene identifier, acquiring the recommendation link corresponding to the scene identifier from the database; determining configuration plug-ins corresponding to the recommended links from a database; and determining the configuration parameters corresponding to the configuration plug-ins from the database.

In one embodiment, a recommendation request implemented by a processor executing a computer program carries a user identification; acquiring configuration information corresponding to the scene identifier, including: determining configuration conditions corresponding to the user identification from at least two groups of configuration conditions of the recommendation link; and determining a configuration plug-in corresponding to the recommendation link according to the configuration condition.

In one embodiment, after obtaining the recommendation result according to the recommendation service, the processor, implemented when executing the computer program, further includes: acquiring a log of the recommended service, wherein the log comprises a recommendation result returned according to the recommended service; analyzing the recommended result in each log to obtain a log analysis result, and adjusting the configuration information according to the log analysis result; and performing recommended service debugging based on the adjusted configuration information.

In one embodiment, the obtaining of the recommendation result from the recommendation service implemented by the processor when executing the computer program comprises: acquiring a current plug-in a current recommendation link; executing the current plug-in to obtain a current intermediate processing result; and acquiring a next plugin in the next recommendation link, taking the next plugin as the current plugin, acquiring a next intermediate processing result based on the current intermediate processing result and the current plugin, taking the next intermediate processing result as the current intermediate processing result, and continuously executing the step of acquiring the next plugin in the next recommendation link until all recommendation links are executed to obtain the recommendation result.

In one embodiment, executing the current plug-in by the processor when executing the computer program results in a current intermediate processing result, comprising: and if the number of the current plug-ins is more than one, performing serial processing or parallel processing on each current plug-in according to the preset identification of each current plug-in to obtain a current intermediate processing result.

In one embodiment, a recommendation service configuration method implemented by a processor executing a computer program comprises: receiving a configured scene identifier; receiving configuration information for a scene identity; storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plugins of the recommendation links and parameters corresponding to the plugins.

In one embodiment, receiving configuration information for a scene identification implemented when the computer program is executed by the processor comprises: receiving at least two groups of configuration conditions of at least one recommended link; and storing at least two groups of configuration conditions in association with the recommended links.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of: receiving a recommendation request, wherein the recommendation request carries a scene identifier; acquiring configuration information corresponding to the scene identifier; generating corresponding recommended service according to the configuration information; and obtaining a recommendation result according to the recommendation service.

In one embodiment, the obtaining configuration information corresponding to a scene identification, which is implemented when the computer program is executed by the processor, includes: calling a configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a database; determining configuration plug-ins corresponding to the recommended links from a database; determining configuration parameters corresponding to the configuration plug-ins from a database; and outputting the recommendation link, the configuration plug-in and the configuration parameters as configuration information.

In one embodiment, a call configuration query service implemented by a computer program when executed by a processor, comprises: calling a configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a cache; if the cache comprises recommendation links corresponding to the scene identification, determining configuration plug-ins corresponding to the recommendation links from the cache; determining configuration parameters corresponding to the configuration plug-ins from the cache; if the cache does not comprise a recommendation link corresponding to the scene identifier, acquiring the recommendation link corresponding to the scene identifier from the database; determining configuration plug-ins corresponding to the recommended links from a database; and determining the configuration parameters corresponding to the configuration plug-ins from the database.

In one embodiment, a recommendation request implemented by a computer program when executed by a processor carries a user identification; acquiring configuration information corresponding to the scene identifier, including: determining configuration conditions corresponding to the user identification from at least two groups of configuration conditions of the recommendation link; and determining a configuration plug-in corresponding to the recommendation link according to the configuration condition.

In one embodiment, the computer program, when executed by the processor, further comprises, after obtaining the recommendation result according to the recommendation service: obtaining a log of the recommended service, wherein the log comprises a recommended result returned according to the recommended service; analyzing the recommended result in each log to obtain a log analysis result, and adjusting the configuration information according to the log analysis result; and performing recommended service debugging based on the adjusted configuration information.

In one embodiment, the computer program, when executed by a processor, implements obtaining recommendations from a recommendation service, comprising: acquiring a current plug-in a current recommendation link; executing the current plug-in to obtain a current intermediate processing result; and obtaining a next plug-in the next recommendation link, taking the next plug-in as the current plug-in, obtaining a next intermediate processing result based on the current intermediate processing result and the current plug-in, taking the next intermediate processing result as the current intermediate processing result, and continuously executing the step of obtaining the next plug-in the next recommendation link until all recommendation links are executed, so as to obtain a recommendation result.

In one embodiment, executing the current plug-in when executed by the processor results in a current intermediate processing result, comprising: and if the number of the current plug-ins is more than one, performing serial processing or parallel processing on each current plug-in according to the preset identification of each current plug-in to obtain a current intermediate processing result.

In one embodiment, a recommendation service configuration method implemented by a computer program when executed by a processor includes: receiving a configured scene identifier; receiving configuration information for a scene identity; storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plugins of the recommendation links and parameters corresponding to the plugins.

In one embodiment, a computer program product is provided, comprising a computer program which when executed by a processor performs the steps of: receiving a recommendation request, wherein the recommendation request carries a scene identifier; acquiring configuration information corresponding to the scene identifier; generating corresponding recommended service according to the configuration information; and obtaining a recommendation result according to the recommendation service.

In one embodiment, the computer program, when executed by the processor, further comprises, after obtaining the recommendation result according to the recommendation service: acquiring a log of the recommended service, wherein the log comprises a recommendation result returned according to the recommended service; analyzing the recommended result in each log to obtain a log analysis result, and adjusting the configuration information according to the log analysis result; and performing recommended service debugging based on the adjusted configuration information.

In one embodiment, the computer program when executed by a processor implements obtaining recommendations according to a recommendation service, comprising: acquiring a current plug-in a current recommendation link; executing the current plug-in to obtain a current intermediate processing result; and obtaining a next plug-in the next recommendation link, taking the next plug-in as the current plug-in, obtaining a next intermediate processing result based on the current intermediate processing result and the current plug-in, taking the next intermediate processing result as the current intermediate processing result, and continuously executing the step of obtaining the next plug-in the next recommendation link until all recommendation links are executed, so as to obtain a recommendation result.

It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the various embodiments provided herein may be, without limitation, general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, or the like.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A recommendation request processing method, the method comprising:

receiving a recommendation request, wherein the recommendation request carries a scene identifier;

acquiring configuration information corresponding to the scene identifier;

generating corresponding recommended service according to the configuration information;

and obtaining a recommendation result according to the recommendation service.

2. The method of claim 1, wherein the obtaining configuration information corresponding to the scene identifier comprises:

calling a configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a database;

determining a configuration plug-in corresponding to each recommended link from the database;

determining configuration parameters corresponding to the configuration plug-ins from the database;

and outputting the recommendation link, the configuration plug-in and the configuration parameters as configuration information.

3. The method of claim 2, wherein invoking the configuration query service comprises:

calling the configuration query service, and acquiring a recommendation link corresponding to the scene identifier from a cache;

if the cache comprises the recommendation links corresponding to the scene identification, determining configuration plug-ins corresponding to the recommendation links from the cache; determining configuration parameters corresponding to the configuration plug-ins from the cache;

if the cache does not comprise the recommendation link corresponding to the scene identifier, acquiring the recommendation link corresponding to the scene identifier from the database; determining a configuration plug-in corresponding to each recommended link from the database; and determining configuration parameters corresponding to each configuration plug-in from the database.

4. The method of claim 2, wherein the recommendation request carries a user identification; the obtaining of the configuration information corresponding to the scene identifier includes:

determining configuration conditions corresponding to the user identification from at least two groups of configuration conditions of the recommended link;

and determining a configuration plug-in corresponding to the recommendation link according to the configuration condition.

5. The method of claim 1, wherein after obtaining the recommendation according to the recommendation service, further comprising:

acquiring a log of a recommended service, wherein the log comprises a recommended result returned according to the recommended service;

analyzing the recommendation result in each log to obtain a log analysis result, and adjusting the configuration information according to the log analysis result;

and performing recommended service debugging based on the adjusted configuration information.

6. The method of claim 1, wherein obtaining the recommendation according to the recommendation service comprises:

acquiring a current plug-in a current recommendation link;

executing the current plug-in to obtain a current intermediate processing result;

and obtaining a next plugin in the next recommendation link, taking the next plugin as a current plugin, obtaining a next intermediate processing result based on the current intermediate processing result and the current plugin, taking the next intermediate processing result as the current intermediate processing result, and continuously executing the step of obtaining the next plugin in the next recommendation link until all recommendation links are executed, so as to obtain a recommendation result.

7. The method of claim 6, wherein executing the current plug-in obtains a current intermediate processing result, comprising:

and if the number of the current plug-ins is more than one, performing serial processing or parallel processing on each current plug-in according to the preset identification of each current plug-in to obtain the current intermediate processing result.

8. A recommendation service configuration method is characterized by comprising the following steps:

receiving a configured scene identifier;

receiving configuration information for the scene identity;

storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plug-ins of the recommendation links and parameters corresponding to the plug-ins.

9. The method of claim 8, wherein the receiving configuration information for the scene identity comprises:

receiving at least two sets of configuration conditions of at least one of the recommended links;

and storing at least two groups of configuration conditions in association with the recommended links.

10. A recommendation request processing apparatus, characterized by comprising:

the system comprises a receiving module, a recommending module and a judging module, wherein the receiving module is used for receiving a recommending request, and the recommending request carries a scene identifier;

the acquisition module is used for acquiring configuration information corresponding to the scene identifier;

the generating module is used for generating corresponding recommended service according to the configuration information;

and the recommendation result obtaining module is used for obtaining a recommendation result according to the recommendation service.

11. A recommendation service configuration apparatus, characterized in that the recommendation service configuration apparatus comprises:

the identification receiving module is used for receiving the configured scene identification;

a configuration receiving module, configured to receive configuration information for the scene identifier;

the storage module is used for storing the scene identification and the configuration information in an associated manner; the configuration information comprises recommendation links, plug-ins of the recommendation links and parameters corresponding to the plug-ins.

12. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7 or 8 to 9.

13. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7 or 8 to 9.

14. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7 or 8 to 9.